Autonomous denudation in an intelligent automated in vitro fertilization and intracytoplasmic sperm injection platform

Abstract

A method for automated, artificial-intelligence-based denudation includes scanning, robotically, a dish containing a cumulus-oocyte-complex (COC) using an imaging system and an artificial intelligence/machine learning system (AI/ML system) to create an image object. The imaging system includes a microscopy system, a camera system, and a lighting system. The dish includes an enzyme to remove cumulus cells from an oocyte among the COC. The method includes identifying the COC within the image object, and a location, based on comparing the image object to a threshold using the AI/ML system. The threshold is an optical pattern with a probability of corresponding to a COC mass. The method includes instructing a robotic pipettor to collect the COC at the location. The method includes iteratively collecting the COC within the robotic pipettor and expelling the COC to return it to the dish until the AI/ML system confirms sufficient removal of cumulus and corona cells.

The invention claimed is: 
     
       1. A method for automated, artificial-intelligence-based denudation, the method comprising:
 scanning, robotically, a dish containing a cumulus-oocyte-complex (COC) using an imaging system and an artificial intelligence/machine learning system (AI/ML system) to create an image object, wherein the imaging system includes at least one of a microscopy system, a camera system, and a lighting system, and wherein the dish includes an enzyme to remove cumulus cells from an oocyte among the COC; 
 identifying the COC within the image object, and a COC location, based on using the AI/ML system to compare the image object with an optical pattern corresponding to a COC mass; 
 instructing a robotic pipettor to collect the COC at the COC location; and 
 iteratively collecting the COC within the robotic pipettor and expelling the COC from the robotic pipettor to return it to the dish until the AI/ML system confirms, using a second image object produced by the imaging system, removal of at least a portion of the cumulus cells from the COC. 
 
     
     
       2. The method of  claim 1  wherein the dish contains a plurality of wells. 
     
     
       3. The method of  claim 1  wherein the dish is a tube. 
     
     
       4. The method of  claim 1  wherein the imaging system includes a stereomicroscope. 
     
     
       5. The method of  claim 1  wherein the imaging system includes an inverted microscope. 
     
     
       6. The method of  claim 1  wherein the imaging system includes a movable microscope. 
     
     
       7. The method of  claim 1  wherein the enzyme is a plurality of enzymes. 
     
     
       8. The method of  claim 1  wherein the enzyme is a plurality of compounds. 
     
     
       9. The method of  claim 1  wherein scanning is automatically performed using the AI/ML system and the imaging system based on a predetermined pattern. 
     
     
       10. The method of  claim 1  wherein scanning is automatically adjusted to use a scanning pattern based on the AI/ML system and the imaging system and an AI/ML system evaluation of at least one output of the imaging system. 
     
     
       11. The method of  claim 1  wherein the robotic pipettor includes a plurality of robotic pipettes. 
     
     
       12. The method of  claim 1  wherein iteratively collecting the COC within the robotic pipettor includes sequentially reducing an internal diameter of pipettes used by the robotic pipettor throughout an iteration. 
     
     
       13. The method of  claim 1  wherein the COC mass is a plurality of COC masses. 
     
     
       14. A method for automated artificial-intelligence-based human egg preparation, the method comprising:
 scanning, robotically, a dish containing a cumulus-oocyte-complex (COC) using an imaging system and an artificial intelligence/machine learning system (AI/ML system) to create a first image object, wherein the imaging system includes at least one of a microscopy system, a camera system, and a lighting system, and wherein the dish includes an enzyme to remove cumulus cells from an oocyte among the COC; 
 identifying the COC within the first image object, and a COC location, based on using the AI/ML system to compare the first image object with an optical pattern corresponding to a COC mass; 
 instructing a first robotic pipettor to collect the COC at the COC location; 
 using the AI/ML system to determine a first intake volume of egg/fluid to the first robotic pipettor; 
 using the AI/ML system and the imaging system to determine a presence of air bubbles within the first intake volume of egg/fluid; 
 discharging the first intake volume of egg/fluid into a receptacle placed on a stage to create a discharged volume of egg/fluid; 
 using the AI/ML system and the imaging system to guide a positioning of a second robotic pipettor to intake a subset volume of egg/fluid from the discharged volume of egg/fluid, wherein the second robotic pipettor is of a narrower internal diameter than the first robotic pipettor; and 
 using a plurality of progressively narrower robotic pipettors, including the second robotic pipettor, to intake and discharge portions of the subset volume of egg/fluid until removal of at least a portion of the cumulus cells is confirmed for at least one egg in the subset volume of egg/fluid by the AI/ML system and the imaging system using a second image object. 
 
     
     
       15. The method of  claim 14  wherein the second robotic pipettor has a same internal diameter as the first robotic pipettor. 
     
     
       16. The method of  claim 14  wherein the stage is an inverted microscope stage. 
     
     
       17. The method of  claim 14  wherein the scanning is automatically performed according to a scan pattern specified by the AI/ML system. 
     
     
       18. The method of  claim 14  wherein the first and second image objects are still images. 
     
     
       19. The method of  claim 14  wherein the first and second image objects are time-series images. 
     
     
       20. The method of  claim 14  wherein the first and second image objects are videos.